Impact of ocean acidification on antimicrobial activity in gills of the blue mussel (Mytilus edulis)

Accepted Manuscript Impact of ocean acidification on antimicrobial activity in gills of the blue mussel (Mytilus edulis) B. Hernroth, S. Baden, H. Tassidis, K. Hörnaeus, J. Guillemant, S. Bergström Lind, J. Bergquist PII:

S1050-4648(16)30157-7

DOI:

10.1016/j.fsi.2016.04.007

Reference:

YFSIM 3917

To appear in:

Fish and Shellfish Immunology

Received Date: 27 January 2016 Revised Date:

7 April 2016

Accepted Date: 8 April 2016

Please cite this article as: Hernroth B, Baden S, Tassidis H, Hörnaeus K, Guillemant J, Bergström Lind S, Bergquist J, Impact of ocean acidification on antimicrobial activity in gills of the blue mussel (Mytilus edulis), Fish and Shellfish Immunology (2016), doi: 10.1016/j.fsi.2016.04.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

pH 8.1 pH 7.7

EP TE D

Vibrio sp.

M AN U

RI PT Intensity

SC

AMPs

Myticin A (average intensity)

AC C

Growth inhibition (% mg-1 mL-1)

ACCEPTED MANUSCRIPT

LPS

m/z

0h 3h

ACCEPTED MANUSCRIPT

1

Impact of ocean acidification on antimicrobial activity in gills of the blue

2

mussel (Mytilus edulis)

3 B. Hernrotha,b,*, S. Badenc, H. Tassidisb, K. Hörnaeusd, J. Guillemantd, S. Bergström Lindd, J.

5

Bergquistd

RI PT

4

6 7

a

8

Kristineberg 566, SE – 451 78 Fiskebäckskil, Sweden.

9

b

10

c

11

566, SE – 451 78 Fiskebäckskil, Sweden.

12

d

13

599, SE – 75124 Uppsala, Sweden.

SC

The Royal Swedish Academy of Sciences, Sven Lovén Center for marine science,

M AN U

Dept. of Natural Science, Kristianstad University, SE – 291 88 Kristianstad, Sweden.

Dept. of Biological and Environmental Sciences, University of Gothenburg, Kristineberg

Dept. of Chemistry – BMC, Analytical Chemistry and SciLifeLab, Uppsala University, Box

TE D

14 15

*Corresponding author: The Royal Swedish Academy of Sciences, Sven Lovén Center for

17

marine science, Kristineberg 566, SE – 451 78 Fiskebäckskil, Sweden.

18

Tel: +46 70 6045120

19

E-mail address: [email protected] (Bodil Hernroth)

AC C

EP

16

1

ACCEPTED MANUSCRIPT

ABSTRACT

21

Here, we aimed to investigate potential effects of ocean acidification on antimicrobial peptide

22

(AMP) activity in the gills of Mytilus edulis, as gills are directly facing seawater and the

23

changing pH (predicted to be reduced from ~8.1 to ~7.7 by 2100). The AMP activity of gill

24

and haemocyte extracts was compared at pH 6.0, 7.7 and 8.1, with a radial diffusion assay

25

against Escherichia coli. The activity of the gill extracts was not affected by pH, while it was

26

significantly reduced with increasing pH in the haemocyte extracts. Gill extracts were also

27

tested against different species of Vibrio (V. parahaemolyticus V. tubiashii, V. splendidus and

28

V. alginoyticus) at pH 7.7 and 8.1. The metabolic activity of the bacteria decreased by ~65-

29

90%, depending on species of bacteria, but was, as in the radial diffusion assay, not affected

30

by pH. The results indicated that AMPs from gills are efficient in a broad pH-range. However,

31

when mussels were pre-exposed for pH 7.7 for four month the gill extracts presented

32

significantly lower inhibit of bacterial growth. A full in-depth proteome investigation of gill

33

extracts, using LC-Orbitrap MS/MS technique, showed that among previously described

34

AMPs from haemocytes of Mytilus, myticin A was found up-regulated in response to

35

lipopolysaccharide, 3 h post injection. Sporadic occurrence of other immune related

36

peptides/proteins also pointed to a rapid response (0.5- 3 h p.i.). Altogether, our results

37

indicate that the gills of blue mussels constitute an important first line defence adapted to act

38

at the pH of seawater. The antimicrobial activity of the gills is however modulated when

39

mussels are under the pressure of ocean acidification, which may give future advantages for

40

invading pathogens.

SC

M AN U

TE D

EP

AC C

41

RI PT

20

42

Keyword

43

Ocean acidification, Mytilus edulis, antimicrobial peptide, gill tissue, Vibrio, LPS, proteome

44

analysis, LC-Orbitrap MS/MS analysis

2

ACCEPTED MANUSCRIPT

Introduction

46

Climate change, caused by increased levels of greenhouse gases, mainly carbon dioxide

47

(pCO2), in the atmosphere, now emerges as one of the most important challenges of the 21st

48

century [1]. Surface water of the world oceans is gradually getting warmer and its uptake of

49

CO2 across the air-sea interface alters seawater carbon chemistry, which is acidifying the

50

oceans. So far, pH has been reduced by 0.1 units, representing an increase in [H+] of about 30

51

% [2]. A further reduction of approximately 0.4 units is predicted by 2100 [3]; a phenomenon

52

termed “ocean acidification” (OA).

SC

RI PT

45

Our recent research has shown that marine invertebrates, such as the sea-star Asterias

54

rubens [4] and the Norway lobsters Nephrops norvegicus [5,6] become immune suppressed

55

when exposed to seawater, mimicking the pH conditions that are predicted to occur by the end

56

of this century. Such OA-effects have recently been shown in the blue mussel, Mytilus edulis,

57

and it was found that the mussels’ bacteriostatic capacity against the bivalve pathogen Vibrio

58

tubiashii was significantly reduced at pH ~7.7 [7].

TE D

59

M AN U

53

The immune defence of mussels is mainly based on encapsulation and phagocytosis by circulating haemocytes [8] including fundamental mechanisms for cell killing, such as

61

reactive oxygen metabolites and lysosomal enzymes [9,10,11]. An acidified environment

62

generates shell dissolution in mussels causing hypercapnia, which may reduce the functional

63

properties of haemocytes [12,13,14]. However, despite shell dissolution the study by Asplund

64

et al. [7] showed that calcium haemostasis was maintained after 4 months of exposure to OA

65

and neither haemocyte counts nor phagocytic capacity were shown affected by the OA-

66

condition. The immunomodulation behind the reduced bacteriostatic capacity may involve

67

other mechanisms such as e.g. transduction of bacterial signals into the host cell, which so far

68

have not been investigated in relation to OA. Neither has potential modulation of the

69

activation of antimicrobial peptides (AMPs) been included in OA-studies, even though they

AC C

EP

60

3

ACCEPTED MANUSCRIPT

constitute important components of mussel immunity [15]. Such broad-spectrum AMPs are

71

potent against a wide range of microbes, including bacteria, fungi, viruses and protozoa and

72

have been described for both plants and animals, as reviewed by e.g. Zasloff [16] and Yeaman

73

and Yount [17].

RI PT

70

The limited access to gene sequences makes proteomic studies of Mytilidae difficult and

75

relatively few proteins have been identified in these non-model organisms [18,19,20]. In the

76

last few decades, six groups of AMPs have been described in the Mediterranean mussel

77

Mytilus galloprovincialis; defensins [21], myticins [22, 23], mytilins [24], mytimicin [25],

78

mytimacins and big defensins [26], and an additional peptide, the myticusin, has been found

79

in M. corusus [27]. Among these, three have been isolated also from haemocytes of M. edulis.

80

These are mytilin (isoform A and B), mytimicin and defensin (isoform A and B) [28,29],

81

which are characterized as cationic, cysteine rich ß-sheets. It has still not been clarified

82

whether AMPs of Mytilus are constitutively expressed [30] or might be inducible [21,28] as in

83

insects and vertebrates.

M AN U

TE D

84

SC

74

Blue mussels are filter-feeding bivalves, and gills are one of the major sites of interaction with the environment. By using gamma camera technique, it has been demonstrated that high

86

concentrations of pathogens can accumulate along the ciliated filaments [31]. In coastal

87

waters the mussels are exposed to pathogens from e.g. improper disposal of human sewage

88

waste and leaching from agricultural activities, including cattle grazing at nearshore

89

meadows. Thus, E. coli are commonly used as fecal indicator bacteria in recreational waters

90

and sea food. However, there are also resident pathogens in seawater, such as bacteria of the

91

genus Vibrio, which are able to infect humans as well as shellfish. Like in other organisms

92

there is a need of an efficient first-line defence on the mussel epithelium. Mitta et al. [32]

93

have shown mytilin immune reactivity in infiltrating haemocytes of the gill epithelium of M.

94

galloprovincialis and Balserio et al. [33] have shown expression of myticin C in granular

AC C

EP

85

4

ACCEPTED MANUSCRIPT

haemocytes infiltrating this tissue. If AMPs are delivered on gill epithelium they are supposed

96

to act at a pH level in accordance to the ambient seawater (pH ~ 8.1), compared to AMPs in

97

haemolymph (pH ~ 7.5) [7] and the acidic vesicles of haemocytes (pH ~ 3-5 as measured in

98

phagolysosomes of oysters by Beaven and Paynter [34]). On a short term, pH of the ambient

99

sea water may fluctuate considerably, particularly in tidal zones, where adapted bivalves can

100

survive for hours by closing the shells during hypoxic events. However, virtually nothing is

101

known about functional properties of epithelium peptides and potential effects of long term

102

exposure to OA.

SC

The aim of this study was to investigate if peptide extracts of the gills of M. edulis have

M AN U

103

RI PT

95

antimicrobial activity against the Gram-negative bacteria E. coli and Vibrio spp. We

105

investigated if the activity was affected by lowered pH, at a level mimicking future ocean

106

acidification (OA). This was carried out through in vitro studies comparing the minimal

107

inhibitory concentration of peptide extracts, from gill tissue and haemocytes of M. edulis,

108

against E. coli at different pH. In another experiment, metabolic activity of different marine

109

bacterial species of the genus Vibrio was investigated after incubation with gill extracts at pH

110

8.1 (pH of seawater today) and 7.7 (pH predicted for 2100). In addition, bacteriostatic

111

capacity of gill extracts from mussels that for four months were pre-exposed to seawater of

112

either pH 8.1 or 7.7, was determined as growth inhibition of V. parahaemolyticus.

113

Furthermore, the full proteome and especially the small proteins, including AMPs of gill

114

extracts were investigated, using LC-Orbitrap MS/MS analysis. Quantitative peptide/protein

115

expression was compared between different time points after immune challenge with

116

lipopolysaccharide (LPS).

AC C

EP

TE D

104

117

5

ACCEPTED MANUSCRIPT

118

2. Material and Methods

119 2.1. Experiment 1: In vitro comparison of antimicrobial activity of peptide extracts at

121

different pH

122

2.1.1. Sampling

123

Blue mussels, M. edulis, (length ~5-7 cm) were collected at ~ 2 m depth in the vicinity of

124

Sven Lovén Centre for Marine Science, Kristineberg, (SLC-Kristineberg) at the Swedish west

125

coast, where the tidal amplitude is only ~ 0.2 m. The mussels were until used, kept in

126

containers supplied with running seawater from the water system at SLC-Kristineberg in a

127

thermo-constant room (~32 PSU, ~ 14 °C).

SC

M AN U

128

RI PT

120

For each sample, haemolymph from ten mussels were pooled. Two mL haemolymph per mussel was squirted into an equal volume of anti-coagulant buffer (0.05 M Tris-HCl pH 7.6,

130

2% glucose, 2% NaCl, 0.5% EDTA) in a 50 mL centrifuge tube and kept on ice. It was

131

centrifuged (Beckman J-18) at 1200 x g at 4°C for 15 min and the pellet was used for

132

extraction of peptides.

133

TE D

129

The gills were dissected and pooled in the same way as for haemocytes, then frozen in 80°C and lyophilised for about 20 h (Lyovac GT 2, Leybold-Heraus) before peptide

135

extraction.

AC C

136

EP

134

137

2.1.2. Peptide extractions

138

The method for extracting peptides from haemocytes was slightly modified after Mitta et al.

139

[22]. Briefly, the haemocyte pellet was homogenized in 50 mM Tris buffer with 50 mM NaCl,

140

pH 8.7. Organelles were collected through centrifugation before sonicated in 2 M ice cold

141

acetic acid. After centrifugation at 10,000 x g at 4°C for 20 min the supernatant was loaded on

142

a Sep-Pak C18 Vac cartridge (Waters Associates) which had been equilibrated with acidified

6

ACCEPTED MANUSCRIPT

(0.05% triflouroacetic acid (TFA); 99%, Acros organics) sterile water, also used to wash out

144

unbound fragments. For eluting the peptides acidified (0.05% TFA) 40 and 80% acetonitrile

145

(CH3CN) were used. Acetonitrile was evaporated in vacuum (SpeedVac Concentrator) before

146

storing the tubes in a freezer until used for the antimicrobial assay.

147

RI PT

143

Gills were homogenised in 10 volumes of 50 mM phosphate buffer, then placed

horizontally on an orbital shaker over night at 6°C before centrifuged (Beckman J-18) at 5320

149

x g at 4°C for 30 min. The supernatant was loaded on a Sep-Pak C-18 cartridge, following the

150

same procedure as described for haemocytes.

Before testing the antimicrobial activity, as described in section 2.1.3, the dry Sep-Pak C18

M AN U

151

SC

148

152

fractions were reconstituted in 100 µL of sterile water and the protein concentrations were

153

determined using Coomassie assay, with bovine serum albumin (BSA) as standard [35].

154

2.1.3. Antimicrobial assay: MIC µg mL-1

156

Minimal inhibitory concentration (MIC µg mL-1) of the peptide fractions, used for the in vitro

157

study, was determined with a radial diffusion assay with double layer agar [36]. The gram-

158

negative bacterium E. coli O14 was cultured in LB-broth (20 g L-1 LB base) adjusted to pH

159

6.0, 7.7 and 8.1, respectively, to mid-logarithmic phase. Hundred µL of bacterial suspensions,

160

at a concentration of 6*104 colony forming units (CFU) µL-1, was added to agarose [0.11 g

161

agarose (DNA-grade), 10 mL 10 mM sodium phosphate buffer, 20 µL LB-broth and 2 µL

162

Tween-20]. The pH was set to 6.0, 7.7 and 8.1, respectively, before used as an under layer in a

163

petri dish (90 mm in diameter). Wells, 3 mm in diameter, were punched in the agar and 5 µL

164

of serial doubling dilutions of the peptide extraction in sterile water were added to the wells.

165

The agar plates were incubated at 37°C for 3 h, before covering the under layer with a

166

nutrient-rich agarose layer with pH adjusted accordingly. The plates were incubated over

167

night at 37°C. A synthetic antimicrobial peptide supplied by Prof. Inger Mattsby-Baltzer

AC C

EP

TE D

155

7

ACCEPTED MANUSCRIPT

(Dept. of Clin. Bact., University of Gothenburg, Sweden) was used as an internal control

169

added to each agarose plate. Sterile water was used as a negative control. The following day

170

zone diameters were measured and MIC-values (µg mL-1) calculated according to Hultmark et

171

al. [37]. The experiment including peptide extraction was repeated 8 times.

172

RI PT

168

2.2. Experiment 2: Comparison of the effect of gill extracts on the metabolic activity of

174

different Vibrio species at different pH

175

2.2.1. Antimicrobial assay: Metabolic index (% mg-1 mL-1)

176

From blue mussels, kept in the running seawater system at SLC-Kristineberg (pH ~ 8.1; 32

177

PSU, 14 °C), gills were dissected and peptides were extracted (n=6) as described in section

178

2.1.2. In this case gill tissue from three individuals were pooled, to obtain a peptide extraction

179

enough to be used for investigating its effect on survival of four different shellfish pathogenic

180

species of the genus Vibrio [38]. These were V. parahemolyticus CCUG 43363, V. tubiashii

181

ATCC 19109, V. splendidus CCUG 20273 and V. alginoyticus CCUG 16315 T. In addition,

182

the antimicrobial activity was tested against E. coli O14 as a reference since it was also used

183

in the MIC-assay (described in section 2.1.3). Stock-cultures of the vibrios were thawed and

184

cultured in 1% Peptone 3.2% NaCl with pH adjusted to 8.1 and 7.7 units, respectively, at RT

185

until stationary phase. Then, 1 mL of the bacteria suspensions was inoculated in a fresh

186

portion of the peptone broth and cultured at RT, until reaching mid-logarithmic phase. E. coli

187

was cultured in the same way but in LB-broth. The bacteria were diluted in PBS 3.2% NaCl

188

(pH 8.1 and 7.7, respectively) to ~ 107 bacteria mL-1 and duplicates of 90 µL of the bacterial

189

suspension were transferred to 96-well microplates. After 1 h incubation at RT, on an orbital

190

shaker, with 10 µL of either gill extract or PBS the metabolic activity, here presented as MI %

191

mg protein mL-1, was measured as dehydrogenase activity, using colorimetric Cell

192

Proliferation Assay (Promega # G5421), according to Hernroth, [39].

AC C

EP

TE D

M AN U

SC

173

8

ACCEPTED MANUSCRIPT

193 2.3 Experiment 3: Antimicrobial activity in gill extracts from mussels pre-exposed to

195

different pH

196

2.3.1. Experimental set up

197

Specimens of M. edulis were collected as described above, and were subsequently maintained

198

in the running seawater system of SLC-Kristineberg (~ 14°C, 32 PSU; pH 8.1 units). Twenty

199

mussels were transferred to and kept for four months (end of November to end of March) in a

200

basket, hanging in each of four 400 L tanks (flow-through: 1 L min-1). For the experimental

201

manipulation a seawater pH predicted to occur by the year 2100 (∆pH ≈ –0.4 units [3] was

202

selected for two of the tanks. This was regulated by manipulation of CO2 levels by adding

203

pure gaseous CO2 directly into the tanks and maintaining pH at 7.7±0.1 using a computerised

204

feed-back system (AquaMedic). Natural seawater without CO2 manipulation was used for the

205

control mussels in the two other tanks (Table 1). Circulating pumps (Eheim 600 submersible

206

pump at a rate of 300 l h-1) secured a stable pH and water renewal for the mussels.

TE D

M AN U

SC

RI PT

194

After four months, gills from the mussels were dissected and immediately frozen in -80°C.

208

Peptide extractions from gill tissue of 6 mussels from each tank were performed as described

209

above, but by in this case using BCA protein detection assay (Thermo Scientific, USA) for

210

determination of protein concentrations.

AC C

211

EP

207

212

2.3.2. Antimicrobial assay: Growth inhibition (% mg-1 mL-1)

213

To compare antimicrobial activity from gill extracts of mussels pre-exposed to ambient

214

seawater (pH 8.1; Control) and CO2 manipulated seawater (pH 7.7; OA) a strain of the

215

bacteria V. parahaemolyticus, was used. This strain has been isolated from M. edulis collected

216

at the same place as the mussels used in this study and has as well been used for infectivity

217

studies on M. edulis in a previous study [40]. Here, gill extracts of single individuals were

9

ACCEPTED MANUSCRIPT

used to compare effects on bacterial growth in broth, using 96 wells microplates. After

219

optimization of the assay the following protocol was established: A stock solution of V.

220

parahaemolyticus was thawed and grown in peptone broth in the same way as described

221

above to reach mid-logarithmic phase. Bacteria were harvested through centrifugation,

222

washed and diluted in peptone broth maintaining pH 7.7 and 8.1, respectively, until OD600

223

was approximately 0.130. Triplicates of 90 µL of the bacterial suspensions were transferred to

224

wells and incubated with 10 µL of either peptide extract or sterile water. After 3 h at 37°C the

225

OD600 was measured in a microplate reader (Victor™ X4, # 2030 Multilabel Reader, Perkin

226

Elmer Inc., Ohio, US) and the capacity of extracts to inhibit bacterial growth was calculated

227

in relation to the OD600 values of bacteria incubated with sterile water. Results were presented

228

as Growth inhibition % mg-1 protein mL-1. Melittin from honey bee venom (#M2272; Sigma

229

Aldrich, USA) was used as positive control. The procedure was repeated 12 times and each

230

time the antimicrobial activity of gills from one OA-treated and one Control-treated mussel

231

was analysed.

SC

M AN U

TE D

232

RI PT

218

2.4. Experiment 4: Peptide expression in gills of mussels

234

2.4.1. Experimental set up

235

In an attempt to quantitatively investigate the expression of AMPs in the gill tissue, 30

236

mussels were kept in the running seawater system of SLC-Kristineberg (~32 PSU, 14°C) and

237

divided into six 15 L basins with five individuals in each. In order to avoid contamination of

238

bacterial proteins the mussel immunity was pre-challenged with lipopolysaccharide (LPS;

239

#L7261, Sigma Aldrich) dissolved in physiological saline (PS) buffer [39]. One control group

240

of mussels was injected with only PS-buffer and the other five groups were injected with 0.2

241

µg LPS g-1 mussel (wet weight), into the adductor muscle. The gills of mussels from the

242

control group were dissected at time 0 followed by dissection of one group at a time after 0.5,

AC C

EP

233

10

ACCEPTED MANUSCRIPT

243

1.5, 3, 5 and 8 h post injection. The dissected gill tissues were immediately put on dry ice

244

before being stored at -80°C until further prepared.

245 2.4.2 Sample preparation

247

The samples were homogenized in 200 µL lysis buffer (9M urea, 20 mM HEPES) using a

248

micro pestle. Proteins were extracted by sonication using a probe with a 3 mm tip (10 x 1 sec,

249

amplitude 30%) followed by centrifugation at 16,000 x g for 20 min at 4°C. The total protein

250

concentration in the samples was analysed as described in section 2.1.2. Aliquots

251

corresponding to 20 µg protein were reduced with dithiothreitol (DTT) and alkylated with

252

iodoacetamide (IAA). After four times dilution with 50 mM ammonium bicarbonate, trypsin

253

was added in a trypsin:protein ratio of 1:20 and digestion was performed overnight.

254

Thereafter the samples were purified by Pierce C18 Spin Columns (Thermo Scientific), dried

255

and resolved in 60 µL 0.1% formic acid.

M AN U

SC

RI PT

246

TE D

256 2.4.3. LC-MS/MS

258

The samples were analysed using a QExactive Plus Orbitrap mass spectrometer (Thermo

259

Fisher Scientific, Bremen, Germany) equipped with a nano electrospray ion source. The

260

peptides were separated by reversed phase liquid chromatography using an EASY-nLC 1000

261

system (Thermo Fisher Scientific). A set-up of pre-column and analytical column was used.

262

The pre-column was a 2 cm EASY-column (1D 100 µm, 5 µm C18) (Thermo Fisher

263

Scientific) while the analytical column was a 10 cm EASY-column (ID 75 µm, 3 µm, C18;

264

Thermo Fisher Scientific). Peptides were eluted with a 90 min linear gradient from 4% to

265

100% acetonitrile at 250 nL min-1. The mass spectrometer was operated in positive ion mode

266

acquiring a survey mass spectrum with resolving power 70,000 (full width half maximum)

AC C

EP

257

11

ACCEPTED MANUSCRIPT

267

and consecutive high collision dissociation (HCD) fragmentation spectra of the 10 most

268

abundant ions.

269 2.5 Data handling and statistical methods

271

2.5.1 Antimicrobial assay

272

When comparing MIC-values from peptide extractions of haemocytes and gills (n=8), eluted

273

with 40 and 80% acetonitrile, respectively, and the activity against E. coli at different pH a

274

three-factor analyses of variances (ANOVA; Holm Sidak test for multiple comparison) was

275

used (factor Tissue, two levels: gills and haemocytes; factor pH, three levels: 8.1, 7.7 and 6.0;

276

factor Elution, two levels 40% and 80%). Two-factor ANOVA (factor Bacterial species, five

277

levels; V. parahaemolyticus, V. tubiashii, V. spledidus, V. alginolyticus, E. coli; factor pH,

278

two levels: 8.1 and 7.7) was used when comparing the metabolic activity between different

279

species of bacteria, exposed to gill extracts at different pH. The activity of gill extracts from

280

mussels that were pre exposed to either ambient (Control) or CO2 manipulated sea water (OA)

281

for four months, before testing the capacity to inhibit growth of V. parahaemolyticus was

282

compared using one factor ANOVA (Student Newman Keuls method for multiple

283

comparison). The data were tested for normal distribution (Shapiro-Wilk) before entering

284

ANOVA and when failed, data were analysed with Kruskal-Wallis One Way Analysis of

285

Variance on Ranks (Tukey Test for multiple comparison). Level of significance was set to

286

p<0.05. The statistical analyses were performed using Sigma Plot, version 12.5 (Jandel

287

Scientific Software, San Rafael, CA).

SC

M AN U

TE D

EP

AC C

288

RI PT

270

289

2.5.2. Peptide/protein identification and quantification

290

The acquired data (.RAW-files) were processed by MaxQuant 1.5.1.2. [41]. Protein

291

identification was performed against a FASTA database containing proteins from Mytilus

12

ACCEPTED MANUSCRIPT

extracted from the NCBI database (release June 2015). A decoy search database, including

293

common contaminants and a reverse database, was used to estimate the identification false

294

discovery rate (FDR). The search parameters included: maximum 10 ppm and 0.6 Da error

295

tolerances for the survey scan and MS/MS analysis, respectively; enzyme specificity was

296

trypsin; maximum one missed cleavage site allowed; cysteine carbamidomethylation was set

297

as static modification and oxidation (M) was set as variable modification. A total label free

298

intensity analysis was performed for each individual sample. Thereafter a biostatistics

299

analysis, comparing the samples from the different time intervals post injection (0.5 h, 1.5, 3,

300

5 and 8 h, respectively) with the control samples (0 h), was performed. Proteins that were

301

present in more than four samples in the respective group were included in the subsequent

302

analysis of differences in expression level between the groups. A two-tailed two sample

303

Student’s t-test was performed between the groups. A p-value <0.05 was considered

304

statistically significant.

M AN U

SC

RI PT

292

306 307

3. Results

TE D

305

3.1 In vitro comparison of antimicrobial activity of peptide extracts at different pH

309

Three-factor ANOVA showed that the antimicrobial activity, determined as MIC µg mL-1, did

310

not depend on the percentage of acetonitrile (40 and 80%) used for eluting the peptides from

311

Sep-Pak C18 Vac cartridge (p=0.560). Thus, the results from the two elution fractions are

312

merged in Fig. 1, showing mean values of MIC (±SEM) µg mL-1. The extracts from both

313

haemocytes and gills showed high antimicrobial activity (as shown by low MIC µg mL-1) at

314

pH 6.0. The multiple comparison test (presented in Fig. 1) showed that the antimicrobial

315

activity of hemocytes decreased significantly at pH 7.7 and even more so at pH 8.1. In the gill

316

extracts the antimicrobial activity was not significantly affected neither at pH 7.7 nor at pH

AC C

EP

308

13

ACCEPTED MANUSCRIPT

317

8.1. The antimicrobial activity of extracts from gills was significantly higher compared to that

318

of haemocytes both at pH 7.7 and 8.1.

319 3.2 Comparison of the effect of gill extracts on the metabolic activity of different Vibrio

321

species at different pH

322

Two-factor ANOVA showed that there was a significant difference in metabolic activity (MI

323

% mg-1 protein ml-1) depending on bacterial species (p<0.001) but not depending on pH

324

(p=0.285). Approximately 35% of V. splendidus (mg-1 mL-1 of the gill extracts), were still

325

metabolically active after one hour of incubation (Fig. 2). The difference was significant in

326

comparison with the other vibrio species [V. tubiashii (Diff of mean 10.2% mg-1 mL-1;

327

p<0.001), V. parahaemolyticus (Diff of mean 21.0% mg-1 mL-1; p<0.001), and V.

328

alginolyticus (Diff of mean 24.6% mg-1 mL-1; p<0.001)]. The activity of the gill extracts

329

against E. coli O14 was as in the MIC-assay (presented in Fig. 1) not affected by pH and the

330

metabolic activity was at the same level as for V. parahaemolyticus and V. tubiashii.

SC

M AN U

TE D

331

RI PT

320

3.3 Bacteriostatic capacity of gill extracts from mussels pre-exposed to different pH

333

As no significant values were observed between data obtained from the analyses of the six gill

334

extracts from the duplicate basins of each of the treatments (pH 8.1 and 7.7, respectively), the

335

samples were regarded as true replicates (n=12). The mean value (±SEM) of growth

336

inhibition (Fig. 3) of V. parahaemolyticus when exposed to peptides extracted from mussels at

337

pH 8.1 was 4.95 ± 0.99 % mg-1 mL-1. This was significantly higher than the growth inhibition

338

at pH 7.7 (0.53 ± 0.30 % mg-1 mL-1; p<0.001).

AC C

EP

332

339 340

3.4 Peptide expression in gills of mussels

341

3.4.1 Peptide expression in gill extracts from LPS challenged mussels

14

ACCEPTED MANUSCRIPT

342

A total number of 202 proteins (protein groups) were identified [42] across the 30 samples

343

from the different time intervals (0, 0.5, 1.5, 3, 5 and 8 h respectively) post injection (p.i.)

344

with LPS. Among the antimicrobial peptides, myticin C was the most abundant (Table 2) and its

RI PT

345

isoform k was also detected at all time intervals, with exception for 8 h. Myticin A was

347

detected in all samples from 0-3 h but occurred less frequently at 5 and 8 h. This AMP was

348

the only one showing significant up-regulation in response to the LPS injection (ratio 0-3 h:

349

1.98; p=0.041). In comparison with the control sample, mytilin B was sporadically detected at

350

all time points, while mytilin A was only detected at one time point (3 h).

M AN U

SC

346

351

Among other proteins of immunological interest, fibrinogen-related protein (gi|312270839;

352

FREP) was significantly up-regulated at 0.5 h (ratio 0.5h/0h 3.8, p=0.015) and 3 h (ratio 3h/0h

353

2.7, p=0.033) compared to the control. CuZn superoxide dismutase (gi|402122771) was

354

down-regulated at 3 h (ratio 3h/0h 0.45, p=0.008) and at 8 h (ratio 8h/0h 0.58, p=0.040). Several immunologically relevant proteins were identified, but could not be statistically

TE D

355

evaluated. The goose-type lysozyme (gi|380860907), bactericidal permeability increasing

357

protein (gi|378748968; BPI) and NF κ transcription factor (gi|444299603) were of special

358

interest since these were not detected in control samples. Heat shock protein (HSP) 90

359

(gi|332356355), sHSP 24.1 (gi|347545633) and heat shock cognate 71 (gi|76780612) were

360

identified by several peptides at all time points. HSP 70 (gi|57635269) and sHSP 22

361

(gi|347545631) were sporadically identified among the time points by one to two peptides.

362

Toll-like receptor (gi|407907629) was detected by one peptide in a single sample at 1.5 h.

363

Apextrin-like protein (gi|339785140) was more equally distributed but only detected in a few

364

samples at all time points. The p38 MAPK protein (gi|545283803) was detected with 2-6

365

peptides in a majority of the samples at all time points. Isoforms of the putative C1q domain

366

protein (gi|325504303; MgC1q) were identified across all samples.

AC C

EP

356

15

ACCEPTED MANUSCRIPT

367 368

4. Discussion

369 Blue mussels are commonly used as bioindicators for human pathogens [43,44,45] as they are

371

sessile filter feeders able to persist in coastal waters with high microbial load. It has been

372

demonstrated that their immune effectors are efficient and rapid inhibitors of microbial

373

growth, protecting them from infectious diseases [46]. However, the immunological functions

374

can be sensitive to diverse environmental stressors [47].

SC

This study demonstrated that the gills of blue mussels (M. edulis) express antimicrobial

M AN U

375

RI PT

370

peptides/proteins (here called AMPs), which are more efficient when acting in the pH of

377

seawater compared to that of AMPs extracted from haemocytes. Obviously, AMPs of

378

haemocytes are adapted to act in an acidic environment, as demonstrated by the decreased

379

antimicrobial activity with increasing pH, while pH did not significantly affect the activity of

380

the gill extracts. This was also shown when incubating gill extracts in vitro at different pH

381

with different species of Vibrio. The incubation lasted only one hour, which was enough to

382

reduce the metabolic activity of the bacteria by ~65-90%. The reduction was not affected by

383

pH (8.1 and 7.7) but depended on the species of bacteria. Vibrio alginolyticus and V.

384

parahaemolyticus were the most sensitive while V. splendidus, which is a significant

385

pathogen in bivalve aquaculture [48], was the most resistant. The two in vitro experiments

386

indicated that AMPs from gills are efficient against both E. coli and Vibrio spp in a broad pH-

387

range. However, in the experiment where the blue mussels were pre-exposed to acidified

388

seawater (pH 7.7) for four months there was an obvious effect of pH. Then, bacteriostatic

389

capacity of the gill extracts against V. parahaemolyticus was found significantly reduced

390

compared to that of extracts from control mussels, kept at pH 8.1. It should be noted that in a

391

recent study this strain of V. parahaemolyticus showed no differences in growth rate, survival

AC C

EP

TE D

376

16

ACCEPTED MANUSCRIPT

and haemolytic properties when cultured in peptone broth at pH 7.7 and 8.1, respectively [6].

393

Thus, we could conclude that the reduced bacteriostatic capacity at the lower pH was neither

394

induced by inferior condition of the bacteria, nor by a direct negative impact of pH on the

395

antimicrobial activity. Rather, the results indicated that the mode and action of antimicrobial

396

compounds in the gills are modulated when mussels are kept under the pressure of ocean

397

acidification. In the footstep of global warming, spread of Vibrio diseases is pointed out as an

398

emergent risk [49], with increasing geographical distribution. If the future climate scenario

399

suppresses the first line defence of mussels, the bacteria may take advantage when invading

400

the host and in that way also increase the risk for transmission to humans.

M AN U

SC

RI PT

392

Among the previously described types of AMPs from M. edulis [28,29] we were in the

402

proteomic study of the gill extracts able to identify mytilin (isoform A and B), mytimicin

403

precursor and defensin (isoform B). In addition, we found peptides showing high similarity to

404

the myticins (isoforms A, B, C) and defensin MGD-1 precursor, previously described in M.

405

galloprovincialis [22]. These have all been isolated from haemocytes and when reported from

406

gill tissues it has been assumed that it was due to haemocyte infiltration [32,33]. The powerful

407

antimicrobial activity of gill extracts that we found at low pH may reinforce this assumption.

408

However, as the activity, in contrast to the activity of haemocyte extracts, did not decrease at

409

the higher pH indicates that there could be additional compounds acting.

EP

Antimicrobial activity has been demonstrated in extracts from tissues constituting the first

AC C

410

TE D

401

411

physical barriers (e.g. labial palps, byssus, mantle and gills) of the horse mussel, Modiolus

412

modiolus [50]. Furthermore, Mercado et al. [51] have shown that a distinct, but still not

413

characterized, class of peptides from gill tissue of M. edulis chilensis had high antimicrobial

414

activity and suggested that these were of epithelial origin. Moreover, transcriptomic mapping

415

of gill tissue of M. galloprovincialis has revealed that constitutive expression of big defensins

416

and macins, such as MgBD3 and mytamacin-2, are not exclusively found in haemocytes [26].

17

ACCEPTED MANUSCRIPT

417

This emphasizes a local AMP production to protect the epithelium, although the vast majority

418

of transcripts of known mussel AMPs originate from stimulated haemocytes [52].

419

To avoid massive migration of haemocytes to the gills, LPS was in the present study distributed through injection into the adductor muscle of the mussels. The Toll-like receptor 4

421

(TLR4) has been suggested as the recognition pathway of LPS in M. edulis, which leads to a

422

signalling cascade, normally resulting in the activation of AMP genes [53]. Three hours post

423

injection, we found a significant up-regulation of myticin A in the gill extracts. Previously, it

424

has been described that myticin A is essentially active against Gram-positive bacteria but has,

425

at relatively high concentrations, also bacteriostatic effect against vibrios [22]. In the gill

426

extracts this AMP was seemingly expressed at a basal level as it was found also in all of the

427

five control samples at time 0 h, but it was quantitatively raised in response to LPS. This in

428

contrast to e.g. mytilin B, which was not detected in any of the control samples and at 3 h

429

only found in three samples. Lack of AMP findings may be due to detection limits but it

430

should not be ruled out that the sporadic occurrence noted here, might reflect a response to

431

LPS. A previous study by Hernroth [39] has shown that the antimicrobial activity of peptide

432

extracts from haemocytes of M. edulis increased and dropped to basic level within 3 h after

433

LPS injection. In addition, in vitro expression of AMP transcripts in haemocytes of M.

434

galloprovincialis reached maximum levels 3 h post-infection [54]. Further findings in the

435

present study, also pointed to a rapid immunological response to LPS, as FREPs significantly

436

increased within 3 hours. In M. edulis, a very diverse set of FREP sequences have been

437

identified, suggesting high capacity to recognize and eliminate different kinds of pathogens

438

[55,56]. Also allowing single peptide identification, some other immune related proteins such

439

as lysozyme, BPI, and NF-kB could be identified after the LPS stimuli but not in control

440

samples. Lysozyme has since long been known as an important part of immunity of bivalves

441

[57], and recently the goose-type lysozyme that we found in the gill extracts, was described in

AC C

EP

TE D

M AN U

SC

RI PT

420

18

ACCEPTED MANUSCRIPT

a study on M. galloprovincialis [58]. The bactericidal permeability increasing protein (BPI)

443

has seen constitutively expressed in epithelium of the oyster Crassostrea gigas, where it plays

444

crucial roles in the innate immune response to Gram-negative bacteria [59]. The kinase-

445

activated transcription factor NF-kB constitutes an important signalling factor in the immune

446

response of bivalves, as reviewed by Canesi et al. [60]. As these immune related AMPs were

447

not detected in the gill extracts at time 0 it may indicate a response to LPS.

RI PT

442

There were several other peptide/proteins detected in the extracts that are involved in

449

immunity of marine bivalves, for example HSP70 [61, 62], complement like domains [63],

450

apextrin-like proteins [64] and p38 MAPK [65]. Despite the lack of statistical evaluation, the

451

diversity of detected immune related peptide/proteins enhances the assumption of a local, first

452

line defence of the gill tissue.

M AN U

453

SC

448

The CuZn superoxide dismutase (SOD) was significantly down-regulated in samples from 3 and 8 h post injection. This was surprising since several inflammatory mechanisms are

455

activated by LPS, releasing reactive oxygen species (ROS), which are target for the anti-

456

oxidant activity of the different SOD isoforms [66]. However, a comparison of expression of

457

CuZn SOD transcripts, between different tissues of M. galloprovincialis, has shown quite

458

moderate expression in gills compared to e.g. haepatopancreas, where the expression was

459

approximately 10 fold higher. Moreover, when following the expression in haepatopancreas

460

and haemocytes after challenged with V. anguillarum the CuZn SOD transcription was up-

461

regulated, but it did not reach significant levels until 72 h and 24 h, respectively [58]. This

462

may explain our results as the sampling only lasted 8 h.

EP

AC C

463

TE D

454

With the limited database available for Mytilus [19] it was not possible to provide full and

464

complete proteome analysis of the gill extracts or to identify potential novel AMPs. However,

465

the proteome study clearly indicated that the gills of mussels express immune related

466

peptide/proteins that should be essential in this organ as first line defence. It is possible that

19

ACCEPTED MANUSCRIPT

oxidative stress caused by hypercapnia and acidosis, following ocean acidification [67,68],

468

disrupts transcription or translation and completion of the antimicrobial components. As the

469

proteome databases are continuously expanding, future studies on any changes of expression

470

of antimicrobial components in the gill tissue of long-term OA-exposed mussels, may be

471

possible.

472

RI PT

467

In summary, the antimicrobial peptides that previously have been isolated from

haemocytes of blue mussels were found in gill tissue. Among those, myticin A was seemingly

474

inducible since it was significantly up-regulated in response to LPS. The AMP expression in

475

gill extracts may derivate from infiltrating haemocytes. However, results of the in vitro

476

antimicrobial assays indicated presence of antimicrobial compounds that are more adapted to

477

act in the pH of seawater, in contrast to those of haemocytes. Furthermore, it was shown that

478

pH did not directly affect the antimicrobial activity of gill extracts but if mussels were pre-

479

exposed to acidified seawater the bacteriostatic capacity was significantly reduced. Weakened

480

first line defence on the gills may cause advantages for invading bacteria in the future climate

481

scenario and could also increase the risk for transmission of pathogens to consumers.

482

TE D

M AN U

SC

473

Acknowledgements

484

This work was supported by the Science for Life Laboratory Mass Spectrometry Based

485

Proteomics Facility in Uppsala. Data storage was obtained and supported by BILS

486

(Bioinformatics Infrastructure for Life Sciences). Financial support from The Royal Society

487

of Arts and Sciences in Gothenburg and O.E. & Edla Johansson foundations (BH), The

488

Swedish Research Council (621-2011-4423 JB), Åke Wiberg, Carl Trygger and Magnus

489

Bergvall foundations (SBL) are acknowledged. Dr. Jia Mi is acknowledged for bioinformatic

490

support regarding relative protein expression. The MSc student Cecilia Larsson (Gothenburg

AC C

EP

483

20

ACCEPTED MANUSCRIPT

491

University) and BSc student Arezoo Mehrara (Kristianstad University) are acknowledged for

492

pilot studies on antimicrobial assays.

493

495 496 497 498 499

References

RI PT

494

IPCC, 2014. Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK/New York, NY, USA.

500 501 502 503 504 505

[2]

Orr JC, Fabry VJ, Aumont O, Bopp L, Doney SC, Feely RA, et al. Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 2005; 437:681-86.

[3]

Caldeira K, Wickett ME. Anthropogenic carbon and ocean pH. Nature 2003; 425:365. doi:10.1038/425365a

506 507 508

[4]

Hernroth B, Baden S, Thorndyke M., Dupont S. Immune suppression of the echinoderm Asterias rubens (L.) following long-term ocean acidification. Aquat Toxicol. 2011; 103:222–24.

509 510 511 512 513 514 515

[5]

Hernroth B, Nilsson Sköld H, Wiklander K, Jutfelt F, Baden S. Stimulated climate change causes immune suppression and protein damage in the crustacean Nephrops norvegicus. Fish & Shellfish Immunol. 2012; 33:1095–101.

[6]

Hernroth B, Krång A-S, Baden S. Bacteriostatic suppression in Norway lobster (Nephrops norvegicus) exposed to manganese or hypoxia under pressure of ocean acidification. Aquat Toxicol. 2015; 159:217-24.

516 517 518 519

[7]

Asplund M, Baden S, Russ S, Ellis R, Ningping Gong, Hernroth, B. Ocean acidification and host-pathogen interactions: blue mussels, Mytilus edulis, encountering Vibrio tubiashii. Env Microbiol. 2013; 16:1029–39; doi:10.1111/14622920.12307.

520 521

[8]

Cheng, TC. The role of lysosomes in molluscan inflammation. Am Zool. 1983; 23:129-44.

522 523

[9]

524 525

[10] Pipe RK. Differential binding of lectins to haemocytes of the mussel Mytilus edulis. Cell Tissue Res. 1990; 261:261-68.

526 527

[11] Pipe RK, Generation of reactive oxygen metabolites by the haemocytes of the mussel, Mytilus edulis, Dev Comp Immunol. 1992; 16:111-22.

528 529

[12] Beesley A, Lowe DM, Pascoe CK, Widdicombe S. Effects of CO2-induced seawater acidification on the health of Mytilus edulis. Clim Res. 2008; 37:215-25.

AC C

EP

TE D

M AN U

SC

[1]

Leippe M, Renwrantz L. Release of cytotoxic and agglutinating molecules by Mytilus haemocytes. Dev Comp Immunol. 1988; 12:297-308.

21

ACCEPTED MANUSCRIPT

[13] Bibby R., Widdicombe S, Parry H, Spicer J, Pipe, R. Effects of ocean acidification on the immune response of the blue mussel Mytilus edulis. Aquat Biol. 2008; 2:67–74.

532 533 534

[14] Matozzo V, Chinellato A, Munari M, Finos L, Bressan M, Marin MG. First evidence of immunomodulation in bivalves under seawater acidification and increased temperature. PLoS ONE 7. 2012; doi:10.1371/journal.pone.0033820.

535 536 537

[15] Wootton EC, Dyrynda EA, Ratcliff NA. Bivalve immunity: comparison between the marine mussel (Mytilus edulis), the edible cockle (Cerastoderma edule) and the razorshell (Ensis siliqua). Fish & Shellfish Immunol. 2003; 15: 195-219.

538 539

[16] Zasloff M. Antimicrobial peptides of multicellular organisms. Nature 2002; 415:38995; doi:10.1038/415389a

540 541 542 543 544

[17] Yeaman MR, Yount NY. Mechanisms of Antimicrobial Peptide Action and Resistance. Pharmacol Rev. 2003; 55:27–55.

545 546 547 548

[19] Campos A, Puerto M, Prieto A, Camean A, Almeida AM, Coelho AV, Vasconcelos V. Protein extraction and two-dimensional gel electrophoresis of proteins in the marine mussel Mytilus galloprovincialis: An important tool for protein expression studies, food quality and safety assessment. J Sci Food Agric. 2013; 93:1779–87.

549 550 551

[20] Suárez-Ulloa V, Fernández-Tajes J, Manfrin C, Gerdol M, Venier P, Eirín-López JM. Bivalve Omics: State of the Art and Potential Applications for the Biomonitoring of Harmful Marine Compounds. Mar Drugs 2013; 11:4370-89; doi:10.3390/md11114370

552 553

[21] Hubert F. A member of the arthropod defensin family from edible Mediterranean mussels (Mytilus galloprovincialis). Euro J Biochem. 1996; 240:302–06.

554 555 556

[22] Mitta G, Hubert F, Noël T, Roch P. Myticin, a novel cysteine-rich antimicrobial peptide isolated from haemocytes and plasma of the mussel Mytilus galloprovincialis. Euro J Biochem. 1999; 265:71–78.

557 558 559 560 561 562 563

[23] Pallavicini A, del Mar Costa M, Gestal C, Dreos R, Figueras A, Venier P, Novoa B. High sequence variability of myticin transcripts in hemocytes of immunestimulated mussels suggests ancient host–pathogen interactions. Dev Comp Immunol. 2008: 32:213–26. [24] Mitta G, Vandenbulcke F, Hubert F, Salzet M, Roch P. Involvement of mytilins in mussel antimicrobial defence. J Biol Chem. 2000; 275:12954-62.

564 565 566

[25] Cantet F, Toubiana M, Parisi M-G, Sonthi M, Cammarata M, Roch P. Individual variability of mytimycin gene expression in mussel. Fish & Shellfish Immunol. 2012; 33:641-44.

567 568 569

[26] Gerdol M, De Moro G, Manfrin C, Venier P, Pallavicini A. Big defensins and mytimacins, new AMP families of the Mediterranean mussel Mytilus galloprovincialis. Dev Comp Immunol. 2012; 36: 390–99

570 571 572

[27] Liao Z, Wang X, Liu H, Fan J, Shen W. Molecular characterization of a novel antimicrobial peptide from Mytilus coruscus. Fish & Shellfish Immunol. 2013; 34:610-16.

SC

RI PT

530 531

AC C

EP

TE D

M AN U

[18] Tomanek L. Proteomics to study adaptations in marine organisms to environmental stress. J Proteomics 2014; 105C:92–106.

22

ACCEPTED MANUSCRIPT

[28] Charlet M, Chernysh S, Philippe H, Hetru C, Hoffmann JA, Bulet P. Innate immunity: isolation of several cysteine-rich antimicrobial peptides from the blood of a mollusc, Mytilus edulis. J Biol Chem. 1996; 271:21808–13.

576 577

[29] Tincu JA, Taylor SW. Antimicrobial Peptides from Marine Invertebrates. Antimicrob Agents Chemother. 2004; 48:3645-54.

578 579

[30] Mitta G, Vandenbulcke F, Roch P. Original involvement of antimicrobial peptides in mussel innate immunity. FEBS Letters 2000; 486: 185-90

580 581 582

[31] Hernroth B, Larsson A, Alpsten M, Edebo L. On the possibility of using radioactive labeling and gamma camera technique to study Salmonella typhimurium in the blue mussel, Mytilus edulis. J Shellfish Res. 2000; 19: 175-81.

583 584 585 586 587 588 589 590

[32] Mitta G, Vandenbulcke F, Noël T, Romestand B, Beauvillain JC, Salzet M, Roch P. Differential distribution and defence involvement of antimicrobial peptides in mussel. J Cell Sci. 2000; 113:2759-69. [33] Balseiro P, Falco´ A, Romero A, Dios S, Martínez-López A, Figueras A, Estepa A, Novoa B. Mytilus galloprovincialis Myticin C: A Chemotactic Molecule with Antiviral Activity and Imunoregulatory Properties. PLoS ONE 2011; 6(8): e23140; doi:10.1371/journal.pone.0023140

591 592

[34] Beaven A E, Paynter KT. Acidification of the phagosomes in Crassostrea virginica hemocytes following engulfment of zymosan. Biol Bull. 1999; 196:26–33.

593 594 595 596 597 598

[35] Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72:248-54. [36] Lehrer RI, Rosenman M, Harwig SSSL, Jackson R, Eisenhauer P. Ultrasensitive assays for endogenous antimicrobial polypeptides. J Immunol Met. 1991; 137: 167-73.

599 600 601

[37] Hultmark D, Engström Å, Andersson K, Steiner H, Bennich H, Boman HG. Insect immunity. Attacins, a family of antibacterial proteins from Hyalophora cecropia. EMBO Journal 1983; 2:571-76.

602 603 604

[38] Romalde JL, Barja JL. Bacteria in molluscs: good and bad guys in A Mendéz-Vilas (Ed). Current research, technology and education topics in applied microbiology and biotechnology. Formatex 2010; 136-47.

605 606 607

[39] Hernroth B. The influence of temperature and dose on antimicrobial peptide response against lipopolysaccharide in the blue mussel, Mytilus edulis. Fish & Shellfish Immunol. 2003; 14:25-37.

608 609 610

[40] Oweson C, Hernroth B. A comparative study on the influence of manganese on the bactericidal response of marine invertebrates. Fish & Shellfish Immunol. 2009; 27:500-07.

611 612 613

[41] Cox J, Mann M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotech. 2008; 26:1367 – 72; doi:10.1038/nbt.1511

614

[42] Hörnaeus K, Guillemant J, Mi J. Hernroth B, Bergquist J, Bergström Lind S Mass

AC C

EP

TE D

M AN U

SC

RI PT

573 574 575

23

ACCEPTED MANUSCRIPT

615 616 617

spectrometry data from a proteomics-based study of antimicrobial activity in gills of the blue mussel (Mytilus edulis). Data in Brief, submitted. [43] Hernroth BE, Conden-Hansson A-C, Rehnstam-Holm A-S, Girones R, Allard AK. Environmental factors influencing human viral pathogens and their potential indicator organisms in the blue mussel, Mytilus edulis: the first Scandinavian report. Appl Env Microbiol. 2002; 68:4523–33.

622 623

[44] Carlos JA, Lees DN. Environmental Transmission of Human Noroviruses in Shellfish Waters. Appl Env Microbiol. 2014; 80:3552–61

624 625 626 627 628 629 630 631 632 633 634 635 636

[45] Dabrowski T, Doré WJ, Lyons K, Nolan GD. Numerical modelling of blue mussel (Mytilus edulis) bacterial contamination. J Sea Res. 2014; 89:52–63.

RI PT

618 619 620 621

SC

[46] Venier P, Varotto L, Rosani U, Millino C, Celegato B, Bernante F, Lanfranchi G, Novoa B, Roch P, Figueras A, Pallavicini A. Insights into the innate immunity of the Mediterranean mussel Mytilus galloprovincialis. BMC Genomics 2011; 12:69.

M AN U

[47] Ellis RP, Parry H, Spicer JI, Hutchinson TH, Pipe RK, Widdicombe S. Immunological function in marine invertebrates: Responses to environmental perturbation. Fish & Shellfish Immunolol. 2011; 30:1209-22. [48] LeRoux F, Gay M, Lambert C, Waechter M, Poubalanne S, Chollet B, Nicolas JL, Berthe F. Comparative analysis of Vibrio splendidus-related strains isolated during Crassostrea gigas mortality events. Aquat. Living Resour. 2002; 15:251–58. [49] Baker-Austin C, Trinanes JA, Taylor NGH, Hartnell R, Siitonen A, Martinez-Urtaza J. Emerging Vibrio risk at high latitudes in response to ocean warming. Nature Clim Change 2012; 3:73–77.

640 641 642

[50] Haug T, Stensvåg K, Olsen ØM, Sandsdalen E, Styrvold OB. Antibacterial activities in various tissues of the horse mussel, Modiolus modiolus. J Inv Pathol. 2004; 85:112– 19.

643 644 645

[51] Mercado L, Schmitt P, Marshall SH, Arenas G. Gill tissues of the mussel Mytilus edulis chilensis: A new source for antimicrobial peptides. Electronic J Biotechnol. ISSN: 0717-3458. 2005; 8:284-90. DOI: 10.2225/vol8-issue3-fulltext-2

646 647 648 649

[52] Moreira R, Balseiro P, Planas JV, Fuste B, Beltran S, Novoa B, Figueras A. Transcriptomics of in vitro immune-stimulated hemocytes from the Manila clam Ruditapes philippinarum using high-throughput sequencing. Plos One 2012; 4: e35009. doi: 10.1371/journal.pone.0035009

650 651 652

[53] Tanguay M, McKenna P, Gauthier-Clerc S, Pellerin J, Danger J-M, Siah A. Sequence analysis of a normalized cDNA library of Mytilus edulis hemocytes exposed to Vibrio splendidus LGP32 strain. Res Immunol. 2013; 3:40–50.

653 654 655 656

[54] Costa MM, Prado-Alvarez M, Gestal C, Roch P, Novoa B, Figueras A. Functional and molecular immune response of Mediterranean mussel (Mytilus galloprovincialis) haemocytes against pathogen-associated molecular patterns and bacteria. Fish & Shellfish Immunol. 2009; 26:515–23.

AC C

EP

TE D

637 638 639

24

ACCEPTED MANUSCRIPT

[55] Zhang H, Wang L, Song L, Song X, Wang B, Mu C. Zhang. A fibrinogen-related protein from bay scallop Argopecten irradians involved in innate immunity as pattern recognition receptor. Fish & Shellfish Immunol. 2009; 26:56-64.

660 661 662 663 664

[56] Romero A, Dios S, Poisa-Beiro L, Costa MM, Posada D, Figueras A, Novoa B. Individual sequence variability and functional activities of fibrinogen-related proteins FREPs) in the Mediterranean mussel (Mytilus galloprovincialis) suggest ancient and complex immune recognition models in invertebrates. Dev Comp Immunol. 2011; 35: 334-44.

665 666

[57] McHenery JG, Birkbeck TH, Allen JA. The occurrence of lysozyme in marine bivalves. Comp Biochem Physiol Part B: Comp Biochem. 1979; 63:25-28.

667 668 669

[58] Wang Q, Yuan Z, Wu H, Liu F, Zhao J. Molecular characterization of a manganese superoxide dismutase and copper/zinc superoxide dismutase from the mussel Mytilus galloprovincialis. Fish & Shellfish Immunol. 2013; 34:1345-51.

670 671 672 673

[59] Schmitt P, Diego Rosa R, Duperthuy M, Lorgeril J, Bachèreand E, DestoumieuxGarzón D. The antimicrobial defense of the Pacific oyster, Crassostrea gigas. How diversity may compensate for scarcity in the regulation of resident/pathogenic microflora. Frontiers Microbiol. 2012; 3;1-17.

674 675

[60] Canesi L, Betti M, Ciacci C, Lorusso LC, Pruzzo C, Gallo G. Cell signalling in the immune response of mussel hemocytes. ISJ 2006; 3:40-49.

676 677 678 679

[61] Song L, Wu L, Ni D, Chang Y, Xu W, Xing K. The cDNA cloning and mRNA expression of heat shock protein 70 gene in the haemocytes of bay scallop (Argopecten irradians, Lamarck 1819) responding to bacteria challenge and naphthalin stress. Fish & Shellfish Immunol. 2006; 21:335-45.

680 681 682 683

[62] Aleng NA, Sung YY, MacRae TH, Wahid MEA. Non-Lethal Heat Shock of the Asian Green Mussel, Perna viridis, Promotes Hsp70 Synthesis, Induces Thermotolerance and Protects Against Vibrio Infection. PlosOne. 2015; doi 10.1371/journal.pone.0135603

684 685 686 687

[63] Gerdol M, Manfrin C, De Moro G, Figueras A, Novoa B, Venier P, Pallavicin A. The C1q domain containing proteins of the Mediterranean mussel Mytilus galloprovincialis: A widespread and diverse family of immune-related molecules. Dev Comp Immunol. 2011; 35: 635–43.

688 689 690 691

[64] Estevez-Calvar N, Romero A, Figueras A, Novoa B. Involvement of pore-forming molecules in immune defense and development of the Mediterranean mussel (Mytilus galloprovincialis). Dev Comp Immunol. 2011; 35:1017-31; doi 0.1016/j.dci.2008.08.005.

692 693 694 695

[65] Canesi L, Betti M, Ciacci C, Scarpato A, Citterio B, Pruzzo C, Gallo G. Signaling pathways involved in the physiological response of mussel hemocytes to bacterial challenge: the role of stress-activated p38 MAP kinases. Dev Comp Immunol 2002; 26:325-34.

696

[66] Halliwell B. Biochemistry of oxidative stress. Biochem Soc Trans. 2007; 35:1147–50.

AC C

EP

TE D

M AN U

SC

RI PT

657 658 659

25

ACCEPTED MANUSCRIPT

[67] Pörtner HO, Langenbuch M, Reipschläger A. Biological Impact of Elevated Ocean CO2 Concentrations: Lessons from Animal Physiology and Earth History. J Oceanogr. 2011; 60:705-18. [68] Tomanek L, Zuzow MJ, Ivanina AV, Beniash E, Sokolova IM. Proteomic response to elevated PCO2 level in eastern oysters, Crassostrea virginica: evidence for oxidative stress. J Exp Biol. 2011; 214:1836-44

RI PT

697 698 699 700 701 702 703 704 705

Figure legends

707

Fig. 1 Mean values (±SEM) of minimal inhibitory concentration (MIC µg mL-1) of peptide

708

extracts from haemocyte and gills when tested against E. coli O14, in a double agar layer

709

assay at pH 6.0, 7.7 and 8.1, respectively (n=8). Different letters above bars show

710

significantly different means, analysed by the multiple comparison test presented below (p <

711

0.05).

M AN U

SC

706

712

Fig. 2. Mean values (±SEM) of metabolic activity (expressed as metabolic index; % per mg

714

protein per ml) of V. parahaemolyticus (Vp), V. tubiashii (Vt), V. splendidus (Vs), V.

715

alginolyticus (Va) and E. coli (Ec) after one hour of exposure to gill peptide extracts in PBS at

716

pH 8.1 and 7.7, respectively (n=6). The activity was determined as dehydrogenase activity

717

and calculated in relation to bacteria incubated only with PBS-buffer at the different pH.

718

Different letters above bars show significantly different means, analyzed by the multiple

719

comparison test (p < 0.05).

EP

AC C

720

TE D

713

721

Fig 3. Mean values (±SEM) of inhibition of growth of V. parahaemolyticus when exposed to

722

peptide extracts of gills from mussels that for four months were pre-exposed to ambient

723

seawater (pH 8.1; control) and CO2-manipulated seawater to mimic the future scenario of

724

ocean acidification (pH 7.7; n=12). The growth was estimated as optic density at 600 nm after

26

ACCEPTED MANUSCRIPT

725

incubation with the extracts in peptone broth, at the different pH, and calculated in relation to

726

bacteria incubated without peptide extracts.

727

*** indicate p<0.001

AC C

EP

TE D

M AN U

SC

RI PT

728

27

ACCEPTED MANUSCRIPT

Table 1. Mean values (± STD) of the experimental parameters in the control (Cont) and ocean acidification (OA) treatments of M. edulis. pHtotal scale at standard temperature and pressure (STP), alkalinity (AT µmol kg−1) and salinity (PSU) were measured values and calculated

saturation (Ω Ar). Treatment

pHtot

Salinity

pCO2

RI PT

values (CO2sys software) were pCO2 (µatm), calcite saturation (Ω Ca) and aragonite

Calcite

AT −1

PSU 31.9 ± 0.14

STP 8.0 ± 0.14

µatm 362 ± 45

µmol kg 2171 ± 57

OA 14°C

31.9 ± 0.14

7.69 ± 0.02

964 ± 63

2134 ± 42

Ω 3.68 ± 0.12

SC

Cont 14°C

AC C

EP

TE D

M AN U

1.61 ± 0.07

Aragonite

Ω 2.34 ± 0.07

1.04 ± 0.02

ACCEPTED MANUSCRIPT

Table 2. Number of samples, out of the 5 samples at each of the time intervals post injection with LPS, in which antimicrobial peptides were detected. 0h

0.5h

1.5h

3h

5h

8h

Myticin A; precursor (gi|6647599)

5

5

5

5*

2

1

Myticin-B; precursor (gi|6647598)

0

1

4

Myticin C (gi|374858625)

5

5

5

Myticin C; isoform k (gi|344692895)

2

4

3

Mytilin-A (gi|6225740)

0

0

0

Mytilin B (gi|6225741)

2

2

Mytimicin; precursor (gi|225638981)

4

5

Defensin B (gi|6225251)

0

1

1

0

2

1

5

5

4

4

1

0

3

0

0

SC

1

4

2

1

2

3

3

1

2

0

0

0

0

0

0

0

0

M AN U

Defensin MGD-1; precursor (gi|51317001)

RI PT

Antimicrobial peptides

* indicates significant up-regulation (p<0.05) compared to that of control (0h) as determined by the quantitative

AC C

EP

TE D

analysis of protein expression, described in section 2.4.4.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Impact of ocean acidification on antimicrobial activity in gills of the blue mussel (Mytilus edulis)

AC C

EP

TE D

M AN U

SC

RI PT

Highlights: • Several immune relevant peptides identified in gill extracts of Mytilus edulis • Myticin A upregulated in the gill extracts in response to LPS • Antimicrobial activity in gills more efficient at pH 8.1 compared to hemocytes • The efficiency reduced after 4 month exposure to ocean acidification (pH 7.7) • Indicating advantages for invading bacteria in the future climate scenario